runtime-llamacpp-v0.1.3

QwenAudio/SenseVoiceruntime-llamacpp-v0.1.3Jun 28, 2026by LauraGPT

AI Summary

A portable rebuild of the llama.cpp runtime binaries with conservative CPU flags to prevent crashes caused by instruction set mismatches. This version disables high-ISA flags to ensure compatibility across different hardware.

Key Highlights

Rebuilds v0.1.2 with conservative CPU flags for generic x64 assets.
Disables ggml native CPU tuning and high-ISA flags (AVX/AVX2/AVX512/VNNI/FMA/F16C/BMI2).
Fixes crashes with `Illegal instruction` (SIGILL) or Windows error `0xC000001D`.
Ensures binaries do not depend on GitHub Actions runner CPU instruction set.

New Features

Portable rebuild
Instruction set compatibility fixes
Conservative CPU flag usage

Full Release Notes

Portable rebuild of the FunASR llama.cpp / GGUF runtime binaries.

This release rebuilds the v0.1.2 runtime with conservative CPU flags for the generic x64 assets. In particular, it disables ggml native CPU tuning and high-ISA AVX/AVX2/AVX512/VNNI/FMA/F16C/BMI2 flags so the prebuilt binaries do not depend on the GitHub Actions runner CPU instruction set.

Use this version if v0.1.2 crashes with `Illegal instruction`, `SIGILL`, or Windows error `0xC000001D`.